123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191 |
- import UnityEngine as ue
- import pandas as pd
- import numpy as np
- import matplotlib.pyplot as plt
- import matplotlib.cm as cm
- import json
- from matplotlib.colors import LinearSegmentedColormap
- from mpl_toolkits.mplot3d import Axes3D
- WIDTH = int(70)
- HEIGHT = int(35)
- INDEX1 = "1"
- INDEX2 = "2"
- INDEX3 = "3"
- COND = "17"
- POSITION_PATH1 = ue.Application.dataPath + '/Data_position/'+ COND +'/Walk'+ INDEX1 +'.csv'
- POSITION_PATH2 = ue.Application.dataPath + '/Data_position/'+ COND +'/Walk'+ INDEX2 +'.csv'
- POSITION_PATH3 = ue.Application.dataPath + '/Data_position/'+ COND +'/Walk'+ INDEX3 +'.csv'
- HEATMAP_PATH = 'Assets/Data_image/'+ COND +'/heatmap3DMultiple.png'
- SAVEFILE = 'Assets/Resources/Json/'+ COND +'/heatmap3DMultiple.json'
- def get_data_hight(path):
- dataset = pd.read_csv(path, sep=';', usecols=["Delta Time", "Position x", "Position z"], decimal=',', dtype={'Delta Time': float, 'Position x': float, 'Position z': float})
- dataset["Position x"] = dataset["Position x"].round(0)
- dataset["Position z"] = dataset["Position z"].round(0)
- dataset["Delta Time"] = dataset["Delta Time"].round(2)
- dataset_human_on_same_spot = dataset.groupby(["Delta Time", "Position x", "Position z"]).size().reset_index(name='AmountOneSpot')
- dataset_human_amount_per_deltaTime = dataset_human_on_same_spot.groupby(['Delta Time']).size().reset_index(name='counts')
-
- solution = np.zeros((HEIGHT, WIDTH))
- old = np.zeros((HEIGHT, WIDTH))
- current = np.zeros((HEIGHT, WIDTH))
- minStart = 0
- for i in range(dataset_human_amount_per_deltaTime.shape[0]):
- for count in range(minStart, minStart + int(dataset_human_amount_per_deltaTime['counts'][i])):
- x = int(dataset_human_on_same_spot['Position x'][count])
- y = int(dataset_human_on_same_spot['Position z'][count])
- if i == 0:
- old[y][x] = dataset_human_on_same_spot['AmountOneSpot'][count]
- else :
- current[y][x] = dataset_human_on_same_spot['AmountOneSpot'][count]
- if i > 0:
- solution = solution + np.absolute(old - current)
- old = current
- current = np.zeros((HEIGHT, WIDTH))
- minStart += int(dataset_human_amount_per_deltaTime['counts'][i])
- return solution
- # 1. Get position data from csv file
- data1 = pd.read_csv(POSITION_PATH1, sep=';', usecols=["Position x", "Position z"], decimal=',', dtype={'Position x': float, 'Position z': float})
- data1 = data1.round(0)
- data2 = pd.read_csv(POSITION_PATH2, sep=';', usecols=["Position x", "Position z"], decimal=',', dtype={'Position x': float, 'Position z': float})
- data2 = data2.round(0)
- data3 = pd.read_csv(POSITION_PATH3, sep=';', usecols=["Position x", "Position z"], decimal=',', dtype={'Position x': float, 'Position z': float})
- data3 = data3.round(0)
- # 2. Group by positions and count appearance
- data_count1 = data1.groupby(['Position x', 'Position z']).size().reset_index(name='counts')
- data_count2 = data2.groupby(['Position x', 'Position z']).size().reset_index(name='counts')
- data_count3 = data3.groupby(['Position x', 'Position z']).size().reset_index(name='counts')
- # 3.1 Assign x, y, z, width, depth, height
- x1 = data_count1["Position x"].tolist()
- y1 = data_count1["Position z"].tolist()
- z1 = np.zeros_like(len(x1))
- # dz1 = data_count1["counts"].tolist() # Change
- x2 = data_count2["Position x"].tolist()
- y2 = data_count2["Position z"].tolist()
- z2 = np.zeros_like(len(x2))
- # dz2 = data_count2["counts"].tolist() # Change
- x3 = data_count3["Position x"].tolist()
- y3 = data_count3["Position z"].tolist()
- z3 = np.zeros_like(len(x3))
- # dz3 = data_count3["counts"].tolist() # Change
- # 3.2 Assign height of the bars
- dz_dynamic1 = get_data_hight(POSITION_PATH1)
- dz1 = []
- for i in range(len(x1)):
- dz1.append(dz_dynamic1[int(y1[i])][int(x1[i])])
- dz_dynamic2 = get_data_hight(POSITION_PATH2)
- dz2 = []
- for i in range(len(x2)):
- dz2.append(dz_dynamic2[int(y2[i])][int(x2[i])])
- dz_dynamic3 = get_data_hight(POSITION_PATH3)
- dz3 = []
- for i in range(len(x3)):
- dz3.append(dz_dynamic3[int(y3[i])][int(x3[i])])
- # 3.3 Add offset to day2 and day3
- x2[:] = [a+0.5 for a in x2[:]]
- x3[:] = [a+0.5 for a in x3[:]]
- y3[:] = [a+0.5 for a in y3[:]]
- # 4. Create figure and axes
- fig = plt.figure()
- ax = fig.add_subplot(111, projection='3d')
- # 5.1 Create custom colormap Day1, Day2, Day3
- cmap1 = LinearSegmentedColormap.from_list(name='day1', colors=[(0.40,0.76,0.65), (0.11,0.62,0.47)])
- cmap2 = LinearSegmentedColormap.from_list(name='day2', colors=[(0.99,0.55,0.38), (0.85,0.37,0.01)])
- cmap3 = LinearSegmentedColormap.from_list(name='day3', colors=[(0.55,0.63,0.80), (0.46,0.44,0.70)])
- # 5.2 Initialize array for coloring the bars
- dz_array1 = np.array(data_count1['counts'])
- fracs1 = dz_array1.astype(float) / dz_array1.max()
- color_values1 = cmap1(fracs1.tolist())
- dz_array2 = np.array(data_count2['counts'])
- fracs2 = dz_array2.astype(float) / dz_array2.max()
- color_values2 = cmap2(fracs2.tolist())
- dz_array3 = np.array(data_count3['counts'])
- fracs3 = dz_array3.astype(float) / dz_array3.max()
- color_values3 = cmap3(fracs3.tolist())
- # 5.3 calculate alpha value of the bars
- alpha_values1 = np.array(data_count1['counts'])
- alpha_values1 = alpha_values1.astype(np.float)
- alpha_values1 = (alpha_values1 - alpha_values1.min()) / (alpha_values1.max() - alpha_values1.min())
- alpha_values2 = np.array(data_count2['counts'])
- alpha_values2 = alpha_values2.astype(np.float)
- alpha_values2 = (alpha_values2 - alpha_values2.min()) / (alpha_values2.max() - alpha_values2.min())
- alpha_values3 = np.array(data_count3['counts'])
- alpha_values3 = alpha_values3.astype(np.float)
- alpha_values3 = (alpha_values3 - alpha_values3.min()) / (alpha_values3.max() - alpha_values3.min())
- # 6. Create the bars
- img = ax.bar3d(x1, y1, z1, 0.5, 0.5, dz1, color=color_values1, shade=False)
- img = ax.bar3d(x2, y2, z2, 0.5, 0.5, dz2, color=color_values2, shade=False)
- img = ax.bar3d(x3, y3, z3, 0.5, 0.5, dz3, color=color_values3, shade=False)
- # 7. Create Colorbar
- color_intense1 = data_count1['counts'].tolist()
- color_map1 = cm.ScalarMappable(cmap=cmap1)
- color_map1.set_array(color_intense1)
- fig.colorbar(color_map1)
- color_intense2 = data_count2['counts'].tolist()
- color_map2 = cm.ScalarMappable(cmap=cmap2)
- color_map2.set_array(color_intense2)
- fig.colorbar(color_map2)
- color_intense3 = data_count3['counts'].tolist()
- color_map3 = cm.ScalarMappable(cmap=cmap3)
- color_map3.set_array(color_intense3)
- fig.colorbar(color_map3)
- # 8.
- ax.set_xlabel('Width')
- ax.set_ylabel('Height')
- ax.set_zlabel('Human dynamic')
- plt.show()
- # 10. Save 3D Heatmap
- # heatmap.get_figure().savefig(HEATMAP_PATH, transparent=True)
- # fig.savefig(HEATMAP_PATH, transparent=True)
- # 11. Save the data from the Heatmap to create it in Unity
- dz1 = (np.array(dz1)/np.array(dz1).max()).tolist()
- dz2 = (np.array(dz2)/np.array(dz2).max()).tolist()
- dz3 = (np.array(dz3)/np.array(dz3).max()).tolist()
- jsonData = {"x1" : x1,
- "y1" : y1,
- "z1" : z1.tolist(),
- "dz1" : dz1,
- "alpha_values1" : alpha_values1.flatten().tolist(),
- "x2" : x2,
- "y2" : y2,
- "z2" : z2.tolist(),
- "dz2" : dz2,
- "alpha_values2" : alpha_values2.flatten().tolist(),
- "x3" : x3,
- "y3" : y3,
- "z3" : z3.tolist(),
- "dz3" : dz3,
- "alpha_values3" : alpha_values3.flatten().tolist()}
- with open(SAVEFILE, 'w') as filehandle:
- json.dump(jsonData, filehandle)
|